The duration of the effects of repeated widespread badger culling on cattle tuberculosis following the cessation of culling

Abstract

Background: In the British Isles, control of cattle tuberculosis (TB) is hindered by persistent infection of wild badger (Meles meles) populations. A large-scale field trial--the Randomised Badger Culling Trial (RBCT)--previously showed that widespread badger culling produced modest reductions in cattle TB incidence during culling, which were offset by elevated TB risks for cattle on adjoining lands. Once culling was halted, beneficial effects inside culling areas increased, while detrimental effects on adjoining lands disappeared. However, a full assessment of the utility of badger culling requires information on the duration of culling effects.

Methodology/principal findings: We monitored cattle TB incidence in and around RBCT areas after culling ended. We found that benefits inside culled areas declined over time, and were no longer detectable by three years post-culling. On adjoining lands, a trend suggesting beneficial effects immediately after the end of culling was insignificant, and disappeared after 18 months post-culling. From completion of the first cull to the loss of detectable effects (an average five-year culling period plus 2.5 years post-culling), cattle TB incidence was 28.7% lower (95% confidence interval [CI] 20.7 to 35.8% lower) inside ten 100 km(2) culled areas than inside ten matched no-culling areas, and comparable (11.7% higher, 95% CI: 13.0% lower to 43.4% higher, p = 0.39) on lands <or=2 km outside culled and no-culling areas. The financial costs of culling an idealized 150 km(2) area would exceed the savings achieved through reduced cattle TB, by factors of 2 to 3.5.

Conclusions/significance: Our findings show that the reductions in cattle TB incidence achieved by repeated badger culling were not sustained in the long term after culling ended and did not offset the financial costs of culling. These results, combined with evaluation of alternative culling methods, suggest that badger culling is unlikely to contribute effectively to the control of cattle TB in Britain.

Figure 1. Estimated effects of proactive culling on the incidence of confirmed cattle TB breakdowns.

Estimates are presented for herds inside trial areas as well as those on adjoining lands ≤2 km outside trial area boundaries. The estimated effects of proactive culling are stratified by time periods defined by the cull dates in the during-trial period, and by 6-month intervals from 1 year after the last proactive cull (the post-trial period).

Figure 2. Extrapolation of overall effects to culling areas of different sizes.

The blue area shows the 95% confidence interval for the overall impact (combining the impact inside the targeted area with that seen ≤2 km² outside) of different sized circular culling areas. The red area shows the impact inside the targeted area only. The estimated overall effect is of increased incidence for areas smaller than 17 km², moving to a decreased incidence when areas larger than 17 km² are targeted. The effect of decreased overall incidence is statistically significant for areas larger than 141 km².